This invention relates to monospecific antibodies which are reactive with a subclass of fibrinogen. More particularly, the invention relates to monospecific antibodies which are reactive with the .alpha..sub.E subunit of this fibrinogen molecule, and to methods of use of such antibodies.
Fibrinogen is one of the more well-studied and abundant proteins in the human circulatory system. Its complex structure--a heavily disulfide-bonded hexamer composed of two copies each of the .alpha., .beta. and .gamma. subunits--and central role in blood clot formation and wound healing account for the high profile it has enjoyed as a subject of both biochemical and medical research. Recently, new attention has been given to structure/function relationships in the fibrinogen molecule. This new interest has in part been prompted by growth in the understanding of this protein's range of activity in normal and pathological states (Refs. 1-3). However, a major impetus to fibrinogen research has been provided by the recent identification of a long overlooked, naturally occurring variant of the a subunit, designated ".alpha..sub.E " (Ref. 4) Unlike the .alpha. subunit, the .alpha..sub.E subunit bears a C-terminal extension which confers significant homology to the .beta. and .gamma. subunits.
Fibrinogen is synthesized and secreted into the circulation by the liver. Circulating fibrinogen is polymerized under attack by thrombin to form fibrin, which is the major component of blood clots or thrombi. Subsequently, fibrin is depolymerized under attack by plasmin to restore the fluidity of the plasma. Many of the steps in the polymerization and depolymerization processes have been well established (Ref. 5). The elevated levels of fibrinogen which are part of the acute phase response occurring in the wake of infections and trauma are now known to come from increased hepatic production, primarily in response to interleukin-6 (IL-6) (Ref. 6).
By the late 1960's , the general subunit structure of fibrinogen was firmly established (Ref. 7) and, a decade later, the complete amino acid sequence was reported (Refs. 8-11). Over the next 10 years, the cluster of three separate genes encoding the .alpha. (alpha), .beta. (beta), and .gamma. (gamma) subunits was identified on chromosome 4q23-q32 (Ref. 12), and the apparently complete genetic sequences of all three fibrinogen subunits were published (Ref. 13). These studies indicated that the .alpha. subunit lacked a globular C-terminal domain comparable to those present in the .beta. and .gamma. subunits.
The subsequent discovery of an additional exon (i.e., exon VI) downstream from the established .alpha. subunit gene has resolved the evolutionary mystery posed by the imperfectly parallel structure of the three major subunits (Refs. 4, 14). A novel fibrinogen .alpha. chain transcript has been identified at low frequency bearing the exon VI-derived sequences as a separate open reading frame. Additional splicing leads to the use of this extra sequence to elongate the .alpha. chain by 35%, providing the subunit with a globular domain (the "VI-domain") similar to those of the .beta. and .gamma. chains. Evidence shows that this previously unidentified extended .alpha. chain (.alpha..sub.E) is assembled into fibrinogen molecules and that its synthesis is enhanced by interleukin-6 (IL-6). These facts suggest that the .alpha..sub.E subunit participates in both the acute phase response and in normal physiology.
Using a polyclonal rabbit antibody preparation specific to the VI-domain, .alpha..sub.E was demonstrated to occur in plasma fibrinogen as part of (.alpha..sub.E .beta..gamma.).sub.2, a homodimeric (i.e., symmetrical) molecule of .about.420 kilodaltons (kDa) (Ref. 15). This species has been designated "Fib.sub.420 " to distinguish it from the abundant 340 kDa form of fibrinogen, denoted "Fib.sub.340 " ((.alpha..beta..gamma.).sub.2). Although the relatively low circulating level of Fib.sub.420 (.about.1% of total fibrinogen) is undoubtedly responsible for its having escaped detection until recently, the two extra globular domains are likely to significantly influence the fibrinogen molecule's multiple binding capacities and functions.
A definitive topology of the fibrinogen molecule awaits resolution of its elusive crystal structure. However, numerous indirect studies (Ref. 5) point to a trinodal structure, in which the amino ends of all of the chains are contained in a central nodule from which two triple chain disulfide-interlocked coiled coils diverge. These coils lead to two distal nodules made up of the globular, carboxy terminal domains of the .beta. and .gamma. subunits. Among the more obvious possible functions of the novel globular extension of the .alpha. chain C-terminus may be protection of the .alpha. chain from proteolytic attack. It is also conceivable that such a domain would alter properties of the protofibrils which constitute the laterally cross-linked fibrin strands, particularly if the domain protruded in a plane distinct from that defined by the other three nodules.
Transcripts encoding fibrinogen subunit counterparts having exceptionally high C-terminal homology to human .alpha..sub.E have been detected thus far in lamprey, where it arises from a second .alpha. gene (Refs. 16,17), as well as in chicken, rabbit, rat, and baboon. This degree of .alpha. subunit-associated globular domain preservation in the vertebrate genome signals an important, if as yet unknown, role for .alpha..sub.E. Clues to the potential significance of variations in the .alpha. chain may lie in the similarity of the extension in .alpha..sub.E, not only to the corresponding regions of the fibrinogen .beta. and .gamma. chains, but also to carboxy domains at the C-termini of a number of non-fibrinogen proteins from fruit fly to man (Refs. 18-25). Where functions are known, these non-fibrinogen proteins are constituents of the extracellular matrix and have adhesive properties. It is expected that continued research will permit the determination of whether the .alpha..sub.E globular domain contributes in a subtle way to the primary function of fibrinogen (clot formation and wound healing) or, following the example of other differentially used exons (Refs. 26-28), promotes an alternative function.
In wound repair, fibrinogen serves as a key protein, achieving rapid arrest of bleeding following vessel injury. It promotes both the aggregation of activated platelets with one another to form a hemostatic plug, as well as endothelial cell binding at the site of injury to seal the margins of the wound. As the most abundant adhesive protein in the blood, fibrinogen attaches specifically to platelets, endothelial cells and neutrophils via different integrins (Ref. 29). Five putative receptor recognition domains on human fibrinogen, distributed over its three subunits, have been identified by in vitro and in vivo analyses (Refs. 30-35). In fibrinogen which contains the variant .alpha..sub.E chains, masking of these sites, as well as addition of new sites, are distinct possibilities with ramifications that must be explored. Molecular tools adequate to this purpose have yet to be developed.
Elevated levels of fibrinogen have been found in patients suffering from clinically overt coronary heart disease, stroke and peripheral vascular disease. Although the underlying mechanisms remain speculative, recent epidemiological studies leave little doubt that plasma fibrinogen levels are an independent cardiovascular risk factor possessing predictive power which is at least as high as that of other accepted risk factors such as smoking, hypertension, hyperlipoproteinemia or diabetes (Refs. 36, 37). The structure of fibrin has been analyzed extensively in vitro (Ref. 5). Only recently, however, has attention been paid to the molecular structure of human thrombi and atherosclerotic plaques with respect to fibrinogen and fibrin products (Ref. 38). Whereas thrombi formed in vivo consist primarily of fibrin II cross-linked by factor XIIIa, fibrinogen itself is a major component of uncomplicated atherosclerotic lesions, particularly fibrous and fatty plaques. Immunohistochemical as well as immunoelectrophoretic analyses indicate that fibrinogen in the aortic intima is comparatively well protected from thrombin and plasmin, and that much of it is deposited through direct cross-linking by tissue transglutaminase without becoming converted to fibrin (Ref. 39). Further understanding of these issues awaits the development of methods for the differential determination of fibrinogen subtypes in medical samples.
Fibrinogen-derived protein is also a major component of the stroma in which tumor cells are embedded, but little is known about its molecular structure. Tumor cells promote the secretion of potent permeability factors which cause leakage of fibrinogen from blood vessels (Ref. 3). Extravascular clotting occurs due to procoagulants associated with tumor cells. The resulting fibrinogen/fibrin matrix is constantly remodeled during tumor growth as a consequence of fibrinolysis induced by tumor cell-derived plasminogen activators. It is assumed that fibrin/fibrinogen degradation products play a role during escape of metastatic tumor cells from the primary tumor. There are indications that integrin .alpha..sub.v .beta..sub.3, which is known to interact with the RGDS site in the C-terminal region of the .alpha. chain, may be an important tumor cell surface receptor since it is preferentially expressed on invasive melanoma (Ref. 40). It is not known what effect the globular domain of Fib.sub.420 's .alpha..sub.E subunit plays in tumor development.
Despite evidence indicating roles for .alpha..sub.E fibrinogen in a variety of physiological processes, it appears that .alpha..sub.E deficiency is not lethal in man. This inference is drawn from a recent report on fibrinogen Marburg, a homozygous case of dysfibrinogenemia (Ref. 41). In the .alpha. gene coding for this abnormal fibrinogen, a single base substitution (A.fwdarw.T) has been identified that changes codon .alpha. 461 AAA (lysine) to TAA (stop). As a result, the carboxy-terminal segment 461 to 625 of the common .alpha. chain is lacking and no formation of .alpha..sub.E is possible. Symptoms displayed by the homozygous propositus consisted of severe hemorrhage after delivery followed by repeated thrombotic events that occurred, paradoxically, despite unusually low fibrinogen levels. It is not clear whether the mutant .alpha. chain itself, or the lack of .alpha..sub.E, is responsible for these symptoms.
Fibrinogen levels increase during normal pregnancy (Ref. 42). There is clinical evidence that supports the hypothesis that fibrinogen and firbrin homeostasis is important in pregnancy: low adult levels of (total) fibrinogen were reported to be associated with spontaneous abortions, while fibrinogen infusion was associated with successful gestation (Refs. 43-45). It is undoubtedly significant that, while the fetal concentration of total fibrinogen at term--as measured in umbilical cord blood plasma--is significantly lower than that of adults, the relative level of the Fib.sub.420 subclass is dramatically (about 10 times) higher. None of these phenomena is understood at the molecular level, bespeaking further need for molecular probes with which to define the role of fibrinogen and its subclasses in the underlying physiological mechanisms.
To this time, no stable, sensitive and precise means has existed for detecting and/or purifying .alpha..sub.E -containing fibrinogen (Fib.sub.420). As noted above, a polyclonal antibody has been generated which exhibits specificity for the .alpha..sub.E subunit (Refs. 4, 15), but such antibodies are notoriously problematic when employed for analytical and diagnostic applications. In particular, polyclonal antibodies by their very nature respond to more than one epitope and, therefore, cannot be used to probe individual subdomains in structure/function analyses of a molecule. Moreover, the specificity of polyclonal antibodies varies from animal to animal, as well as with every immunization, as the various antibody subpopulations fluctuate. Indeed, it is not uncommon that only a single animal can be found which is responsive to an immunogen. These problems prohibit the development of precise, accurate and reproducible methods, tests and diagnostics involving the specific identification of .alpha..sub.E subunit of fibrinogen.
As a result, there exists a need for highly specific, sensitive and reproducible probes for enhancing the understanding of the structure and function of fibrinogen, especially in relation to the .alpha..sub.E subunit thereof. There also exists a need for probes suitable for the detection and purification of the .alpha..sub.E subunit and fibrinogen incorporating the subunit. In addition, means for diagnostic testing of subjects with respect to the amount and distribution of fibrinogen in the body are needed. The present invention effectively addresses these and other needs for the first time.